CN111402135A - Image processing method, image processing device, electronic equipment and computer readable storage medium - Google Patents

Abstract

The embodiment of the application provides an image processing method, an image processing device, a mobile terminal and a computer readable storage medium. The method comprises the following steps: carrying out face recognition on an image to be collected; when the image to be collected comprises a face area, collecting a plurality of frames of first images according to a plurality of groups of exposure parameters; performing fusion processing on the multiple frames of first images to obtain a second image; extracting a portrait area contained in the first image, and performing enhancement processing on the portrait area, wherein the portrait area comprises the face area; and splicing the enhanced portrait area with the second image to obtain a target image. The image processing method, the image processing device, the mobile terminal and the computer readable storage medium can enable the person image to show more image details and improve the imaging quality of the person image.

Description

Translated fromChinese

图像处理方法、装置、电子设备及计算机可读存储介质Image processing method, apparatus, electronic device, and computer-readable storage medium

技术领域technical field

本申请涉及计算机技术领域，具体涉及一种图像处理方法、装置、移动终端及计算机可读存储介质。The present application relates to the field of computer technologies, and in particular, to an image processing method, an apparatus, a mobile terminal, and a computer-readable storage medium.

背景技术Background technique

随着电子科技技术的快速发展，用户对使用电子产品进行拍摄的图片质量要求越来越高，而拍摄的图片质量往往受限于拍摄环境的环境亮度、拍摄角度等条件。在拍摄环境的条件较差的情况下拍摄的人物图片无法清晰地呈现更多的细节，成像质量较低。With the rapid development of electronic technology, users have higher and higher requirements for the quality of pictures taken with electronic products, and the quality of pictures taken is often limited by conditions such as ambient brightness and shooting angle of the shooting environment. The pictures of people taken under the conditions of the shooting environment are not able to clearly present more details, and the image quality is lower.

发明内容SUMMARY OF THE INVENTION

本申请实施例公开了一种图像处理方法、装置、移动终端及计算机可读存储介质，能够使人物图像展示更多的图像细节，提高人物图像的成像质量。The embodiments of the present application disclose an image processing method, a device, a mobile terminal, and a computer-readable storage medium, which can make a person image show more image details and improve the imaging quality of the person image.

本申请实施例提供一种图像处理方法，包括：对待采集图像进行人脸识别；当所述待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像；对所述多帧第一图像进行融合处理，得到第二图像；提取所述第一图像中包含的人像区域，并对所述人像区域进行增强处理，所述人像区域包括所述人脸区域；将增强处理后的人像区域与所述第二图像进行拼接，得到目标图像。An embodiment of the present application provides an image processing method, including: performing face recognition on an image to be collected; when the image to be collected includes a face region, collecting multiple frames of first images according to multiple sets of exposure parameters; Frame the first image for fusion processing to obtain a second image; extract the portrait area included in the first image, and perform enhancement processing on the portrait area, where the portrait area includes the human face area; The target image is obtained by splicing the portrait area with the second image.

本申请实施例提供一种图像处理装置，包括：人脸识别模块，用于对待采集图像进行人脸识别；采集模块，用于当所述待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像；融合模块，用于对所述多帧第一图像进行融合处理，得到第二图像；增强模块，用于提取所述第一图像中包含的人像区域，并对所述人像区域进行增强处理，所述人像区域包括所述人脸区域；拼接模块，用于将增强处理后的人像区域与所述第二图像进行拼接，得到目标图像。An embodiment of the present application provides an image processing device, including: a face recognition module for performing face recognition on an image to be collected; a collection module for, when the to-be-collected image includes a face area, The parameters collect multiple frames of the first image; the fusion module is used to fuse the multiple frames of the first image to obtain a second image; the enhancement module is used to extract the portrait area included in the first image, and analyze the The portrait area is enhanced, and the portrait area includes the human face area; a splicing module is used for splicing the enhanced portrait area and the second image to obtain a target image.

本申请实施例提供一种电子设备，包括存储器及处理器，所述存储器中存储有计算机程序，所述计算机程序被所述处理器执行时，使得所述处理器实现如上所述的方法。An embodiment of the present application provides an electronic device, including a memory and a processor, where a computer program is stored in the memory, and when the computer program is executed by the processor, the processor implements the above method.

本申请实施例提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如上所述的方法。Embodiments of the present application provide a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented.

上述实施例公开的图像处理方法、装置、移动终端及计算机可读存储介质，通过对待采集图像进行人脸识别，当待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像，对该多帧第一图像进行融合处理，得到第二图像，提取第一图像中包含的人像区域，并对人像区域进行增强处理，再将增强处理后的人像区域与第二图像进行拼接，得到目标图像，单独对人像区域进行增强处理后再与由多帧第一图像融合得到的第二图像拼接，在保证整体图像的成像效果的同时，也增强人像的视觉效果，能够使人物图像展示更多的图像细节，提高人物图像的成像质量。The image processing method, device, mobile terminal and computer-readable storage medium disclosed in the above embodiments perform face recognition on the image to be collected, and when the image to be collected includes a face area, collect multiple frames of the first image according to multiple sets of exposure parameters. image, perform fusion processing on the multiple frames of the first image to obtain a second image, extract the portrait area included in the first image, perform enhancement processing on the portrait area, and then splicing the enhanced portrait area with the second image , obtain the target image, enhance the portrait area separately, and then splicing it with the second image obtained by merging multiple frames of the first image, while ensuring the imaging effect of the overall image, it also enhances the visual effect of the portrait. Show more image details and improve the imaging quality of human images.

附图说明Description of drawings

为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

图1为一个实施例中图像处理电路的框图；1 is a block diagram of an image processing circuit in one embodiment;

图2为一个实施例中图像处理方法的流程图；2 is a flowchart of an image processing method in one embodiment;

图3为一个实施例中对采集的多帧第一图像进行融合得到第二图像的示意图；3 is a schematic diagram of obtaining a second image by fusing multiple frames of first images collected in one embodiment;

图4为一个实施例中提取第一图像中的人脸区域进行增强处理后再与第二图像进行拼接的示意图；4 is a schematic diagram of extracting a face region in a first image for enhancement processing and then splicing with a second image in one embodiment;

图5为另一个实施例中图像处理方法的流程图；5 is a flowchart of an image processing method in another embodiment;

图6为一个实施例中选取曝光补偿值小于0的第一图像的流程图；6 is a flowchart of selecting a first image whose exposure compensation value is less than 0 in one embodiment;

图7为另一个实施例中图像处理方法的流程图；7 is a flowchart of an image processing method in another embodiment;

图8为另一个实施例中图像处理方法的流程示意图；8 is a schematic flowchart of an image processing method in another embodiment;

图9为一个实施例中图像处理装置的框图；9 is a block diagram of an image processing apparatus in one embodiment;

图10为一个实施例中电子设备的结构框图。FIG. 10 is a structural block diagram of an electronic device in one embodiment.

具体实施方式Detailed ways

下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

需要说明的是，本申请实施例及附图中的术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。It should be noted that the terms "comprising" and "having" in the embodiments of the present application and the accompanying drawings and any modifications thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes unlisted steps or units, or optionally also includes For other steps or units inherent to these processes, methods, products or devices.

可以理解，本申请所使用的术语“第一”、“第二”等可在本文中用于描述各种元件，但这些元件不受这些术语限制。这些术语仅用于将第一个元件与另一个元件区分。举例来说，在不脱离本申请的范围的情况下，可以将第一客户端称为第二客户端，且类似地，可将第二客户端称为第一客户端。第一客户端和第二客户端两者都是客户端，但其不是同一客户端。It will be understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first client may be referred to as a second client, and similarly, a second client may be referred to as a first client, without departing from the scope of this application. Both the first client and the second client are clients, but they are not the same client.

本申请实施例提供一种电子设备。该电子设备中包括图像处理电路，图像处理电路可以利用硬件和/或软件组件实现，可包括定义ISP(Image Signal Processing，图像信号处理)管线的各种处理单元。图1为一个实施例中图像处理电路的框图。为便于说明，图1仅示出与本申请实施例相关的图像处理技术的各个方面。Embodiments of the present application provide an electronic device. The electronic device includes an image processing circuit, and the image processing circuit may be implemented by hardware and/or software components, and may include various processing units that define an ISP (Image Signal Processing, image signal processing) pipeline. FIG. 1 is a block diagram of an image processing circuit in one embodiment. For the convenience of description, FIG. 1 only shows various aspects of the image processing technology related to the embodiments of the present application.

如图1所示，图像处理电路包括ISP处理器140和控制逻辑器150。成像设备110捕捉的图像数据首先由ISP处理器140处理，ISP处理器140对图像数据进行分析以捕捉可用于确定成像设备110的一个或多个控制参数的图像统计信息。成像设备110可包括一个或多个透镜112和图像传感器114。图像传感器114可包括色彩滤镜阵列(如Bayer滤镜)，图像传感器114可获取每个成像像素捕捉的光强度和波长信息，并提供可由ISP处理器140处理的一组原始图像数据。姿态传感器120(如三轴陀螺仪、霍尔传感器、加速度计等)可基于姿态传感器120接口类型把采集的图像处理的参数(如防抖参数)提供给ISP处理器140。姿态传感器120接口可以采用SMIA(Standard Mobile Imaging Architecture，标准移动成像架构)接口、其它串行或并行摄像头接口或上述接口的组合。As shown in FIG. 1 , the image processing circuit includes anISP processor 140 and acontrol logic 150 . Image data captured byimaging device 110 is first processed byISP processor 140 , which analyzes the image data to capture image statistics that can be used to determine one or more control parameters ofimaging device 110 .Imaging device 110 may include one ormore lenses 112 andimage sensor 114 .Image sensor 114 , which may include an array of color filters (eg, Bayer filters), may obtain light intensity and wavelength information captured by each imaging pixel and provide a set of raw image data that may be processed byISP processor 140 . The attitude sensor 120 (eg, a three-axis gyroscope, Hall sensor, accelerometer, etc.) may provide the acquired image processing parameters (eg, anti-shake parameters) to theISP processor 140 based on the interface type of theattitude sensor 120 . The interface of theattitude sensor 120 may adopt an SMIA (Standard Mobile Imaging Architecture, Standard Mobile Imaging Architecture) interface, other serial or parallel camera interfaces, or a combination of the above interfaces.

此外，图像传感器114也可将原始图像数据发送给姿态传感器120，姿态传感器120可基于姿态传感器120接口类型把原始图像数据提供给ISP处理器140，或者姿态传感器120将原始图像数据存储到图像存储器130中。In addition,image sensor 114 may also send raw image data to gesturesensor 120, which may provide raw image data toISP processor 140 based on thegesture sensor 120 interface type, orgesture sensor 120 may store the raw image data to animage memory 130 in.

ISP处理器140按多种格式逐个像素地处理原始图像数据。例如，每个图像像素可具有8、10、12或14比特的位深度，ISP处理器140可对原始图像数据进行一个或多个图像处理操作、收集关于图像数据的统计信息。其中，图像处理操作可按相同或不同的位深度精度进行。TheISP processor 140 processes raw image data pixel by pixel in various formats. For example, each image pixel may have a bit depth of 8, 10, 12, or 14 bits, and theISP processor 140 may perform one or more image processing operations on the raw image data, collecting statistical information about the image data. Among them, the image processing operations can be performed with the same or different bit depth precision.

ISP处理器140还可从图像存储器130接收图像数据。例如，姿态传感器120接口将原始图像数据发送给图像存储器130，图像存储器130中的原始图像数据再提供给ISP处理器140以供处理。图像存储器130可为存储器装置的一部分、存储设备、或电子设备内的独立的专用存储器，并可包括DMA(DirectMemory Access，直接直接存储器存取)特征。TheISP processor 140 may also receive image data from theimage memory 130 . For example, thegesture sensor 120 interface sends the raw image data to theimage memory 130, and the raw image data in theimage memory 130 is provided to theISP processor 140 for processing. Theimage memory 130 may be a part of a memory device, a storage device, or an independent dedicated memory in an electronic device, and may include a DMA (Direct Memory Access, direct memory access) feature.

当接收到来自图像传感器114接口或来自姿态传感器120接口或来自图像存储器130的原始图像数据时，ISP处理器140可进行一个或多个图像处理操作，如时域滤波。处理后的图像数据可发送给图像存储器130，以便在被显示之前进行另外的处理。ISP处理器140从图像存储器130接收处理数据，并对该处理数据进行原始域中以及RGB和YCbCr颜色空间中的图像数据处理。ISP处理器140处理后的图像数据可输出给显示器160，以供用户观看和/或由图形引擎或GPU(Graphics Processing Unit，图形处理器)进一步处理。此外，ISP处理器140的输出还可发送给图像存储器130，且显示器160可从图像存储器130读取图像数据。在一个实施例中，图像存储器130可被配置为实现一个或多个帧缓冲器。When receiving raw image data from theimage sensor 114 interface or from theattitude sensor 120 interface or from theimage memory 130, theISP processor 140 may perform one or more image processing operations, such as temporal filtering. The processed image data may be sent toimage memory 130 for additional processing before being displayed. TheISP processor 140 receives processed data from theimage memory 130 and performs image data processing in the original domain and in the RGB and YCbCr color spaces. The image data processed by theISP processor 140 may be output to thedisplay 160 for viewing by a user and/or further processed by a graphics engine or a GPU (Graphics Processing Unit, graphics processor). In addition, the output of theISP processor 140 may also be sent to theimage memory 130 , and thedisplay 160 may read image data from theimage memory 130 . In one embodiment,image memory 130 may be configured to implement one or more frame buffers.

ISP处理器140确定的统计数据可发送给控制逻辑器150。例如，统计数据可包括陀螺仪的振动频率、自动曝光、自动白平衡、自动聚焦、闪烁检测、黑电平补偿、透镜112阴影校正等图像传感器114统计信息。控制逻辑器150可包括执行一个或多个例程(如固件)的处理器和/或微控制器，一个或多个例程可根据接收的统计数据，确定成像设备110的控制参数及ISP处理器140的控制参数。例如，成像设备110的控制参数可包括姿态传感器120控制参数(例如增益、曝光控制的积分时间、防抖参数等)、照相机闪光控制参数、照相机防抖位移参数、透镜112控制参数(例如聚焦或变焦用焦距)或这些参数的组合。ISP控制参数可包括用于自动白平衡和颜色调整(例如，在RGB处理期间)的增益水平和色彩校正矩阵，以及透镜112阴影校正参数。Statistics determined byISP processor 140 may be sent to controllogic 150 . For example, the statistics may include gyroscope vibration frequency, auto exposure, auto white balance, auto focus, flicker detection, black level compensation,lens 112 shading correction, etc.image sensor 114 statistics.Control logic 150 may include a processor and/or microcontroller executing one or more routines (eg, firmware) that may determine control parameters and ISP processing ofimaging device 110 based on received statistics control parameters of thecontroller 140. For example, control parameters ofimaging device 110 may includeattitude sensor 120 control parameters (eg, gain, integration time for exposure control, stabilization parameters, etc.), camera flash control parameters, camera stabilization displacement parameters,lens 112 control parameters (eg, focus or zoom with focal length) or a combination of these parameters. ISP control parameters may include gain levels and color correction matrices for automatic white balance and color adjustment (eg, during RGB processing), andlens 112 shading correction parameters.

在一个实施例中，通过成像设备(摄像头)110中的透镜112和图像传感器114获取待采集图像，并将待采集图像发送至ISP处理器140。ISP处理器140对待采集图像进行人脸识别，当待采集图像中包含人脸区域时，可获取多组曝光参数，并将多组曝光参数发送至控制逻辑器150。控制逻辑器150可根据多组曝光参数控制成像设备110采集多帧第一图像，并将采集的多帧第一图像发送至ISP处理器140。ISP处理器140接收成像设备110采集的多帧第一图像，对多帧第一图像进行融合处理，得到第二图像，并提取第一图像中包含的人像区域，对该人像区域进行增强处理，再将增强处理后的人像区域与第二图像进行拼接，得到目标图像。In one embodiment, the to-be-captured image is acquired through thelens 112 and theimage sensor 114 in the imaging device (camera) 110 , and the to-be-captured image is sent to theISP processor 140 . TheISP processor 140 performs face recognition on the to-be-captured image, and when the to-be-captured image includes a face region, it can acquire multiple sets of exposure parameters and send the multiple sets of exposure parameters to thecontrol logic 150 . Thecontrol logic 150 may control theimaging device 110 to collect multiple frames of first images according to multiple sets of exposure parameters, and send the collected multiple frames of first images to theISP processor 140 . TheISP processor 140 receives multiple frames of the first images collected by theimaging device 110, performs fusion processing on the multiple frames of the first images to obtain a second image, and extracts a portrait area included in the first image, and performs enhancement processing on the portrait area, Then, the enhanced portrait area is spliced with the second image to obtain a target image.

在一些实施例中，ISP处理器140可将目标图像发送至图像存储器130进行存储。ISP处理器140也可将目标图像输出至显示器160，以供用户对目标图像进行观看等。In some embodiments, theISP processor 140 may send the target image to theimage memory 130 for storage. TheISP processor 140 may also output the target image to thedisplay 160 for the user to view the target image and the like.

如图1所示，在一个实施例中，提供一种图像处理方法，该图像处理方法可适用于手机、智能穿戴设备、平板电脑、数码相机等电子设备，本申请实施例不做限定。其中，上述的电子设备的操作系统可包括但不限于Android操作系统、IOS操作系统、Symbian(塞班)操作系统、Windows操作系统等，本申请实施例不做限定。该方法包括以下步骤：As shown in FIG. 1 , in one embodiment, an image processing method is provided, and the image processing method is applicable to electronic devices such as mobile phones, smart wearable devices, tablet computers, digital cameras, etc., which are not limited in the embodiments of the present application. Wherein, the operating system of the electronic device may include, but is not limited to, an Android operating system, an IOS operating system, a Symbian (Symbian) operating system, a Windows operating system, and the like, which are not limited in the embodiments of the present application. The method includes the following steps:

步骤210，对待采集图像进行人脸识别。Step 210, performing face recognition on the image to be collected.

电子设备可通过摄像头等成像设备获取待采集图像，待采集图像可以指的是通过成像设备实时捕获的预览画面，预览画面仅提供对当前所需拍摄的画面的预览功能，而没有进行成像处理并存储在存储器中。待采集图像可以理解为成像设备的取景画面。The electronic device can obtain images to be captured through imaging devices such as cameras. The images to be captured can refer to the preview images captured in real time by the imaging devices. The preview images only provide the preview function of the currently captured images without performing imaging processing and processing. stored in memory. The image to be captured can be understood as a viewfinder image of the imaging device.

可对待采集图像进行人脸识别，检测待采集图像中是否包含有人脸，待采集图像中可能包含有一张或多张人脸，也可能不包含有人脸。在一些实施方式中，可采用人脸检测算法或人脸检测模型对待采集图像进行人脸识别，例如，可采用VJ(Viola-Jones)算法、MTCNN(多任务级联卷积神经网络)算法等进行人脸检测。Face recognition can be performed on the image to be collected to detect whether the image to be collected contains a human face. The image to be collected may contain one or more faces, or may not contain a human face. In some embodiments, a face detection algorithm or a face detection model can be used to perform face recognition on the image to be collected, for example, a VJ (Viola-Jones) algorithm, an MTCNN (Multi-Task Cascade Convolutional Neural Network) algorithm, etc. can be used. Perform face detection.

在一个实施例中，可基于待采集图像的图像特征判断待采集图像是否包含有人脸。可将待采集图像输入预先构建的人脸检测模型中，人脸检测模型可提取待采集图像中的图像特征。该图像特征可包括但不限于边缘特征、角点特征、颜色特征等，其中，边缘特征用于描述图像中的边缘情况，边缘特征可包括人脸的轮廓特征等，角点特征可指的是图像中存在的拐角或是深度变化大的特征点，比如可包括人脸的眼角特征点、嘴角特征点等，颜色特征则用于描述图像中各个区域的颜色及亮度等。可对提取的图像特征进行分析，判断该图像特征是否与预设的人脸图像特征匹配，若匹配，则可确定待采集图像中包含有人脸。In one embodiment, it may be determined whether the image to be captured contains a human face based on image features of the image to be captured. The image to be collected can be input into a pre-built face detection model, and the face detection model can extract image features in the image to be collected. The image features may include, but are not limited to, edge features, corner features, color features, etc., wherein the edge features are used to describe the edge situation in the image, and the edge features may include contour features of a face, etc. The corner features may refer to The corners in the image or the feature points with large depth changes, such as the feature points of the corners of the eyes, the corners of the mouth, etc. of the face, and the color features are used to describe the color and brightness of each area in the image. The extracted image features can be analyzed to determine whether the image features match the preset face image features, and if so, it can be determined that the image to be collected contains a human face.

在一个实施例中，将待采集图像输入预先构建的人脸检测模型后，可将待采集图像划分为多个小窗口，再分别检测每个小窗口中是否包含有人脸。上述的人脸检测模型均可利用大量的人脸图像及非人脸图像作为样本进行训练学习。In one embodiment, after inputting the image to be collected into a pre-built face detection model, the image to be collected may be divided into multiple small windows, and then it is detected whether each small window contains a human face. The above-mentioned face detection models can use a large number of face images and non-face images as samples for training and learning.

可选地，人脸检测模型可直接输出待采集图像中是否包含人脸的结果，也可以输出待采集图像包含人脸的概率及不包含人脸的概率。当人脸的概率与不包含人脸的概率的差值大于预设值时，若人脸的概率较大，则确定待采集图像中包含有人脸，若非人脸的概率较大，则可确定待采集图像中不包含有人脸。当人脸的概率与不包含人脸的概率的差值小于预设值时，可说明二者概率较为接近，可重新获取待采集图像进行检测。Optionally, the face detection model can directly output the result of whether the image to be collected contains a face, and can also output the probability that the image to be collected contains a face and the probability that the image does not contain a face. When the difference between the probability of a human face and the probability of not containing a human face is greater than the preset value, if the probability of a human face is large, it is determined that the image to be collected contains a human face, and if the probability of not containing a human face is large, it can be determined The images to be collected do not contain human faces. When the difference between the probability of the human face and the probability of not including the human face is smaller than the preset value, it can be shown that the probabilities of the two are relatively close, and the to-be-collected image can be re-acquired for detection.

在一些实施例中，用户可在待采集图像中选取感兴趣区域(region ofinterest，ROI)，感兴趣区域可指的是用户在进行图像拍摄时比较关注的区域。由于在进行图像拍摄时，若拍摄画面中存在人，则用户的感兴趣区域大部分在人脸上。因此，可获取待采集图像被选取的感兴趣区域，并检测感兴趣区域中是否包含人脸，可减少人脸检测的计算量，提高运算速度。In some embodiments, the user may select a region of interest (ROI) in the image to be captured, and the region of interest may refer to a region that the user pays more attention to when capturing images. Because during image capturing, if there is a person in the captured image, most of the user's area of interest is on the face of the person. Therefore, the region of interest selected by the image to be collected can be acquired, and whether the region of interest contains a human face can be detected, which can reduce the amount of calculation of face detection and improve the operation speed.

在一些实施方式中，除了检测待采集图像中是否包含有人脸以外，还可获取检测到的人脸在待采集图像中的位置，确定待采集图像中的人脸区域。可选地，待采集图像中的人脸区域可以采用人脸在待采集图像中的像素坐标进行表示。In some embodiments, in addition to detecting whether the image to be collected contains a human face, the position of the detected face in the image to be collected can also be obtained to determine the face area in the image to be collected. Optionally, the face area in the image to be collected may be represented by the pixel coordinates of the face in the image to be collected.

步骤220，当待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像。Step 220: When the image to be collected includes a face region, collect multiple frames of first images according to multiple sets of exposure parameters.

当确定待采集图像中包含人脸时，可获取多组曝光参数。曝光参数可包括但不限于曝光补偿值、实际曝光量及基础曝光量等，其中，基础曝光量可指的是摄像头的测光模块采集的测光数据对应的曝光量，曝光补偿值则是指相对基础曝光量进行增加或减少的曝光量，实际曝光量则指的是拍摄图像时的真实曝光量。When it is determined that the image to be collected contains a human face, multiple sets of exposure parameters can be obtained. Exposure parameters may include, but are not limited to, exposure compensation value, actual exposure amount, and basic exposure amount. The exposure amount that is increased or decreased relative to the base exposure amount, and the actual exposure amount refers to the actual exposure amount when the image is captured.

在一个实施例中，实际曝光量可根据基础曝光量及曝光补偿值进行确定，例如，曝光补偿值为EV+1，其中，EV指的是摄像头的测光模块采集的测光数据对应的曝光量(即基础曝光量)与实际曝光量的差值，“+”表示增加曝光，数值“1”表示的是补偿曝光的级别，EV+1即指的是相对基础曝光量增强一档曝光，则实际曝光量可以为基础曝光量的两倍。In one embodiment, the actual exposure amount may be determined according to the basic exposure amount and the exposure compensation value, for example, the exposure compensation value is EV+1, where EV refers to the exposure corresponding to the light metering data collected by the light metering module of the camera The difference between the exposure amount (that is, the basic exposure amount) and the actual exposure amount, "+" means to increase the exposure, the value "1" indicates the level of compensation exposure, EV+1 means to increase the exposure by one step relative to the basic exposure amount, Then the actual exposure can be twice the base exposure.

电子设备可根据多组不同的曝光参数对待采集图像进行采集，得到多帧第一图像，每帧第一图像可分别对应一组曝光参数。作为一种具体实施方式，可根据多组不同的曝光参数进行曝光控制，控制摄像头以不同的光圈值和/或快门速度采集得到多帧第一图像，每帧第一图像的图像亮度及包含的细节信息等可不同。可选地，采集的第一图像的帧数可以为3帧，可分别对应具有减少曝光量的曝光补偿值(EV-)、具有不调整曝光量的曝光补偿值、具有增加曝光量的曝光补偿值(EV+)的3组不同的曝光参数。采集的第一图像的帧数也可以是5帧、6帧、8帧等，在此不作限定。The electronic device may collect the images to be collected according to multiple sets of different exposure parameters, and obtain multiple frames of first images, and each frame of the first image may correspond to a set of exposure parameters respectively. As a specific implementation, exposure control can be performed according to multiple sets of different exposure parameters, and the camera is controlled to collect multiple frames of first images with different aperture values and/or shutter speeds. Details and the like may vary. Optionally, the number of frames of the captured first image may be 3 frames, which may correspond to an exposure compensation value (EV-) with reduced exposure, an exposure compensation value with no exposure adjustment, and an exposure compensation with increased exposure. Value (EV+) of 3 groups of different exposure parameters. The number of frames of the collected first image may also be 5 frames, 6 frames, 8 frames, etc., which is not limited here.

步骤230，对多帧第一图像进行融合处理，得到第二图像。Step 230: Perform fusion processing on multiple frames of the first image to obtain a second image.

可将采集的多帧具有不同图像信息的第一图像进行融合，可得到含有更多图像信息的第二图像。进行融合的方式可以为多种，如可利用灰度加权平均法、逻辑滤波法等对多帧第一图像中的相匹配的像素进行融合，也可提取每帧第一图像的图像特征，并对提取的图像特征进行融合得到第二图像等，本申请实施例不进行限定。The collected multiple frames of first images with different image information can be fused to obtain a second image containing more image information. There are many ways to perform fusion, for example, gray-scale weighted average method, logical filtering method, etc. can be used to fuse matching pixels in multiple frames of first images, or the image features of each frame of first image can be extracted, and The extracted image features are fused to obtain a second image, etc., which is not limited in this embodiment of the present application.

在一个实施例中，可采用高动态范围成像(High Dynamic Range Imaging，HDR)对多帧分别对应不同曝光参数的第一图像进行合成，可得到拥有更多动态范围(亮度范围)和图像细节的第二图像。在利用HDR算法进行图像融合时，可根据每帧第一图像包含的图像信息确定每帧第一图像对应的权重，并根据每帧第一图像对应的权重对多帧第一图像进行合成，得到第二图像。第二图像中每个像素点的像素值，可根据每帧第一图像中相匹配的像素点及对应的权重进行加权和计算得到。In one embodiment, high dynamic range imaging (High Dynamic Range Imaging, HDR) may be used to synthesize multiple frames of first images corresponding to different exposure parameters, so as to obtain an image with more dynamic range (brightness range) and image details. second image. When using the HDR algorithm for image fusion, the weight corresponding to the first image of each frame can be determined according to the image information contained in the first image of each frame, and multiple frames of the first image can be synthesized according to the weight corresponding to the first image of each frame. second image. The pixel value of each pixel in the second image can be obtained by weighted sum calculation according to the matched pixel in each frame of the first image and the corresponding weight.

作为一种具体的实施方式，可将第一图像划分成多个区域，每个区域中包含的像素点可以为像素信息较为接近的像素点，例如，可以是深度值较为接近的像素点，或是亮度值、颜色值较为接近的像素点等。可将每帧第一图像中相匹配的区域进行比较，并根据每帧第一图像中相匹配的区域所包含的图像细节的丰富程度，确定每帧第一图像中各个区域对应的权重。可选地，区域的图像细节的丰富程度可根据该区域中包含的像素点邻域的梯度、亮度等进行确定。在进行图像合成时，第二图像中每个像素点的像素值，可根据每帧第一图像中相匹配的像素点的像素值，以及该相匹配的像素点所属的区域对应的权重进行加权和计算得到。可使得合成得到的第二图像包含的图像信息更丰富。As a specific implementation manner, the first image may be divided into multiple regions, and the pixels included in each region may be pixels with closer pixel information, for example, may be pixels with closer depth values, or It is a pixel with a relatively close brightness value and color value, etc. The matched areas in each frame of the first image may be compared, and the weights corresponding to each area in each frame of the first image may be determined according to the richness of image details contained in the matched areas in each frame of the first image. Optionally, the richness of the image details of the area may be determined according to the gradient, brightness, etc. of the neighborhood of the pixel points contained in the area. During image synthesis, the pixel value of each pixel in the second image can be weighted according to the pixel value of the matched pixel in each frame of the first image and the corresponding weight of the region to which the matched pixel belongs and calculated. The image information contained in the second image obtained by synthesis can be enriched.

在一个实施例中，电子设备还可进行运动检测，检测在采集多帧第一图像的过程中是否发生运动。作为一种具体实施方式，可获取成像设备在采集多帧第一图像时的姿态传感器的姿态数据，根据该姿态数据判断是否是发生运动。若多帧第一图像对应的姿态数据发生变化，则可说明发生运动。作为另一种具体实施方式，还可直接利用采集的多帧第一图像判断是否发生运动。可提取少量每帧第一图像中的特征点，并将提取的特征点进行匹配，如果各帧第一图像中相匹配的特征点的像素坐标出现不同，则可说明发生运动。可以理解地，检测是否发生运动的方式可以有多种，并不仅限于上述描述的几种方式。In one embodiment, the electronic device may also perform motion detection to detect whether motion occurs during the process of acquiring multiple frames of the first image. As a specific implementation manner, attitude data of the attitude sensor when the imaging device collects multiple frames of the first image can be acquired, and whether motion has occurred is determined according to the attitude data. If the posture data corresponding to the first images of the multiple frames changes, it can be indicated that movement occurs. As another specific implementation manner, it is also possible to directly use the collected first images of multiple frames to determine whether motion occurs. A small number of feature points in the first image of each frame can be extracted, and the extracted feature points can be matched. If the pixel coordinates of the matched feature points in the first images of each frame are different, it can indicate that motion occurs. It can be understood that there may be various ways to detect whether motion occurs, and it is not limited to the above-described ways.

当检测到在采集多帧第一图像的过程中发生运动时，在对多帧第一图像进行合成前，还可先将多帧第一图像进行配准，使得每帧第一图像上的每个像素点匹配，可使得后续合成的第二图像更为准确。When it is detected that motion occurs in the process of collecting the first images of multiple frames, before synthesizing the first images of the multiple frames, the first images of the multiple frames can also be registered, so that each frame of the first image of each frame can be registered. The matching of pixel points can make the second image synthesized subsequently more accurate.

利用HDR对多帧第一图像进行融合处理，可得到拥有更多动态范围(亮度范围)和图像细节的第二图像，可提高最终成像的成像质量。Using HDR to fuse multiple frames of the first image, a second image with more dynamic range (brightness range) and image details can be obtained, which can improve the image quality of the final image.

步骤240，提取第一图像中包含的人像区域，并对人像区域进行增强处理，人像区域包括人脸区域。Step 240 , extracting a portrait area included in the first image, and performing enhancement processing on the portrait area, where the portrait area includes a human face area.

电子设备采集多帧第一图像后，可从采集的多帧第一图像中选取至少一帧第一图像，并提取该选取的第一图像中包含的人像区域，将人像区域从第一图像中分割出来。可选地，可按照预设的选取策略选择第一图像进行人像区域提取，选取策略可根据实际需求进行设定，例如，可以选取图像质量高的第一图像，图像质量可以根据第一图像中包含的噪声数量及细节信息等进行确定，也可以选取实际曝光量低于基础曝光量的第一图像，或是选取图像亮度在一定范围内的第一图像等，在此不作限定。选取的第一图像的噪声可较少，其人像区域包含的细节信息可较多。After the electronic device collects multiple frames of the first image, it can select at least one frame of the first image from the multiple frames of the collected first image, extract the portrait area included in the selected first image, and extract the portrait area from the first image. split out. Optionally, the first image can be selected to extract the portrait region according to a preset selection strategy, and the selection strategy can be set according to actual needs. For example, the first image with high image quality can be selected, and the image quality can be determined according to the The amount of noise and detailed information contained therein are determined, and the first image with the actual exposure lower than the basic exposure, or the first image with the image brightness within a certain range, etc. may also be selected, which is not limited here. The selected first image may have less noise, and the portrait area thereof may contain more detailed information.

在一些实施方式中，可采用图像分割算法提取第一图像包含的人像区域。可先根据图像的灰度、颜色、纹理及形状等特征将该第一图像划分成若干个互不相交的区域，同一区域内的特征可表现出一致性或相似性，不同区域之间的特征表现出明显不同。划分区域后，再判断各个区域是否属于人像区域，作为一种实施方式，可将各个区域的特征与预设的人像区域特征进行比较，判断是否匹配，若匹配，则可确定该区域属于人像区域。例如，当检测到区域的颜色特征与人像的皮肤颜色特征相同或相似时，则可确定该区域属于人像区域；当检测到区域的纹理特征与人像的衣物纹理特征相似时，则可确定该区域属于人像区域。In some embodiments, an image segmentation algorithm can be used to extract the portrait region included in the first image. The first image can be divided into several disjoint regions according to the characteristics of the image such as grayscale, color, texture and shape. The features in the same region can show consistency or similarity, and the features between different regions showed a marked difference. After dividing the areas, it is then judged whether each area belongs to a portrait area. As an embodiment, the characteristics of each area can be compared with the preset characteristics of the portrait area to determine whether they match. If they match, it can be determined that the area belongs to the portrait area. . For example, when the color feature of the detected area is the same or similar to the skin color feature of the portrait, it can be determined that the area belongs to the portrait area; when the texture feature of the detected area is similar to the clothing texture feature of the portrait, it can be determined that the area is Belongs to the portrait area.

在一些实施方式中，还可根据待采集图像的人脸检测结果确定第一图像中的人脸区域，并根据该人脸区域辅助提取人像区域。可以人脸区域为基准，逐一判断在人脸区域的相邻位置的相邻像素点是否符合人像特征，若符合，则将该相邻像素点划分到人像区域内，并继续检测该被划分到人像区域内的像素点的相邻像素点是否符合人像特征，以此类推，直至找不到符合人像特征的相邻像素点为止。以人脸区域为出发点，进行区域生长，最终得到完整的人像区域，可使提取的人像区域更为准确。In some embodiments, the face region in the first image can also be determined according to the face detection result of the image to be collected, and the face region can be assisted in extracting the face region according to the face region. The face area can be used as the benchmark to judge whether the adjacent pixels in the adjacent positions of the face area conform to the characteristics of the portrait. If so, the adjacent pixels are divided into the portrait area, and the detection is continued Whether the adjacent pixels of the pixels in the portrait area conform to the portrait feature, and so on, until no adjacent pixels conforming to the portrait feature are found. Taking the face region as the starting point, the region grows, and finally a complete portrait region is obtained, which can make the extracted portrait region more accurate.

可以理解地，也可以采集其它方式提取人像区域，并不仅限于上述描述的几种方式。It can be understood that other ways can also be collected to extract the portrait region, and it is not limited to the above-described ways.

电子设备将人像区域从第一图像中分割出来后，可对提取的人像区域进行增强处理，以使增强处理后的人像区域具有更多的图像信息，提高人像区域的视觉效果。在一个实施例中，对人像区域进行增强处理可包括提高人像区域的清晰度、对人像区域进行细节增强等，例如，可提高人像区域的分辨率、对人像区域的五官进行增强等，但不限于此。After segmenting the portrait region from the first image, the electronic device can perform enhancement processing on the extracted portrait region, so that the enhanced portrait region has more image information and improves the visual effect of the portrait region. In one embodiment, the enhancement processing for the portrait area may include improving the definition of the portrait area, enhancing the details of the portrait area, etc. For example, the resolution of the portrait area may be improved, the facial features of the portrait area may be enhanced, etc., but not limited to this.

在一些实施方式中，可基于各种图像增强算法对人像区域进行增强处理，例如，以对人像区域的五官进行增强为例，可采用高通滤波及梯度法等算法锐化五官的边缘轮廓，以突出五官信息，在此不进行限定。In some implementations, the portrait region can be enhanced based on various image enhancement algorithms. For example, taking the enhancement of the facial features of the portrait region as an example, algorithms such as high-pass filtering and gradient methods can be used to sharpen the edge contours of the facial features to enhance Highlight the facial features information, which is not limited here.

可以理解地，步骤230及步骤240并不限制先后执行顺序，可以先执行步骤230，再执行步骤240，也可以先执行步骤240，再执行步骤230，或是步骤230及步骤240并行同时执行。It can be understood thatsteps 230 and 240 are not limited in the order of execution.Steps 230 and 240 may be executed first, then steps 240 may be executed first, and then steps 230 may be executed, orsteps 230 and 240 may be executed concurrently.

步骤250，将增强处理后的人像区域与第二图像进行拼接，得到目标图像。Step 250, splicing the enhanced portrait region with the second image to obtain a target image.

电子设备对从第一图像提取的人像区域进行增强处理后，可将增强处理后的人像区域与第二图像进行拼接，得到无缝的目标图像，该目标图像可指的是成像图像，可用于输出到显示器中进行显示，或是输出到图像处理器中作进一步的图像处理等操作。After the electronic device performs enhancement processing on the portrait region extracted from the first image, the enhanced portrait region can be spliced with the second image to obtain a seamless target image. The target image may refer to an imaging image and can be used for Output to the display for display, or output to the image processor for further image processing and other operations.

在一些实施例中，可先将增强处理后的人像区域与第二图像进行配准，将增强处理后的人像区域的各个像素点与第二图像的像素点进行匹配。可选地，可先确定第二图像中的人脸区域，将第二图像中的人脸区域与增强处理后的人像区域的人脸区域进行配准后，以配准的人脸区域为基准，进行增强处理后的整个人像区域与第二图像的匹配，可使配准的结果更为准确。In some embodiments, the enhanced portrait region may be registered with the second image first, and each pixel of the enhanced portrait region may be matched with pixels of the second image. Optionally, the face region in the second image can be determined first, and after the face region in the second image is registered with the face region of the enhanced portrait region, the registered face region is used as a benchmark. , and the matching of the entire portrait region after the enhancement processing with the second image can make the registration result more accurate.

在完成配准后，可将增强处理后的人像区域与第二图像融合，从而得到目标图像。作为一种具体实施方式，进行图像融合时，可直接用增强处理后的人像区域的像素点替换第二图像中匹配的像素点，也可以为增强处理后的人像区域的像素点及第二图像中匹配的像素点分配不同的权重，再基于分配的权重进行融合，在此不进行限定。After the registration is completed, the enhanced portrait region can be fused with the second image to obtain the target image. As a specific implementation manner, when performing image fusion, the pixels of the enhanced portrait area can be directly replaced with the pixels matched in the second image, or the pixels of the enhanced portrait area and the second image can be replaced. Different weights are assigned to the matched pixels in the , and then fused based on the assigned weights, which is not limited here.

在一些实施方式中，进行图像融合时，可为增强处理后的人像区域的边缘像素点及第二图像中匹配的像素点分配不同的权重，基于分配的权重进行融合，而对于非边缘像素点，可直接用增强处理后的人像区域的像素点替换第二图像中匹配的像素点，在保证人像区域拥有更多图像信息的同时，可使得融合的边缘更为自然，提高视觉效果。In some embodiments, when performing image fusion, different weights may be assigned to the edge pixels of the enhanced portrait region and the matched pixels in the second image, and the fusion is performed based on the assigned weights, while for non-edge pixels , the matched pixels in the second image can be directly replaced by the pixels of the enhanced portrait area, which can make the fused edge more natural and improve the visual effect while ensuring that the portrait area has more image information.

图3为一个实施例中对采集的多帧第一图像进行融合得到第二图像的示意图。图4为一个实施例中提取第一图像中的人脸区域进行增强处理后再与第二图像进行拼接的示意图。如图3所示，电子设备可根据3组不同的曝光参数分别采集得到3帧第一图像(a)、第一图像(b)及第一图像(c),可对采集的3帧第一图像(a)、第一图像(b)及第一图像(c)进行融合处理，得到第二图像310。相比起第一图像(a)、第一图像(b)及第一图像(c)，第二图像310拥有更多动态范围。如图4所示，可选取第一图像(a)，并可对第一图像(a)进行图像切割处理，提取人像区域402，再对人像区域402进行增强处理，得到增强处理后的人像区域404。增强处理后的人像区域404具有更丰富的图像细节。可将增强处理后的人像区域404与第二图像310进行拼接，得到目标图像410，既保证了目标图像410拥有更多动态范围，也增强了人像的视觉效果。可以理解地，图3及图4仅用于说明本申请实施例，并不表示本申请实施例提供的方法在实际应用中的真实成像，不对本申请实施例的成像效果进行限定。FIG. 3 is a schematic diagram of obtaining a second image by fusing multiple frames of first images collected in an embodiment. FIG. 4 is a schematic diagram of extracting a face region in a first image, performing enhancement processing, and then splicing it with a second image in an embodiment. As shown in Figure 3, the electronic device can acquire three frames of the first image (a), the first image (b) and the first image (c) respectively according to three sets of different exposure parameters. The image (a), the first image (b) and the first image (c) are fused to obtain asecond image 310 . Thesecond image 310 has more dynamic range than the first image (a), the first image (b), and the first image (c). As shown in FIG. 4 , the first image (a) can be selected, and image cutting processing can be performed on the first image (a) to extract theportrait area 402, and then theportrait area 402 can be enhanced to obtain the enhanced portrait area. 404. Theenhanced portrait region 404 has richer image details. Theenhanced portrait area 404 and thesecond image 310 can be spliced to obtain thetarget image 410, which not only ensures that thetarget image 410 has more dynamic range, but also enhances the visual effect of the portrait. It is understandable that FIG. 3 and FIG. 4 are only used to illustrate the embodiments of the present application, and do not represent real imaging of the methods provided in the embodiments of the present application in practical applications, and do not limit the imaging effects of the embodiments of the present application.

在本申请实施例中，通过对待采集图像进行人脸识别，当待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像，对该多帧第一图像进行融合处理，得到第二图像，提取第一图像中包含的人像区域，并对人像区域进行增强处理，再将增强处理后的人像区域与第二图像进行拼接，得到目标图像，单独对人像区域进行增强处理后再与由多帧第一图像融合得到的第二图像拼接，在保证整体图像的成像效果的同时，也增强人像的视觉效果，能够使人物图像展示更多的图像细节，提高人物图像的成像质量。In the embodiment of the present application, by performing face recognition on the image to be collected, when the image to be collected includes a face area, collect multiple frames of first images according to multiple sets of exposure parameters, and perform fusion processing on the multiple frames of first images, Obtain the second image, extract the portrait area contained in the first image, perform enhancement processing on the portrait area, and then splicing the enhanced portrait area with the second image to obtain the target image, and enhance the portrait area separately. It is then spliced with the second image obtained by merging multiple frames of the first image, which not only ensures the imaging effect of the overall image, but also enhances the visual effect of the portrait, which can make the portrait image show more image details and improve the imaging quality of the portrait image. .

如图5所示，在一个实施例中，提供另一种图像处理方法，可应用于上述的电子设备。该方法可包括如下步骤：As shown in FIG. 5 , in one embodiment, another image processing method is provided, which can be applied to the above-mentioned electronic device. The method may include the steps of:

步骤502，检测当前拍摄场景。Step 502, detecting the current shooting scene.

电子设备可检测当前的拍摄场景，拍摄场景可根据拍摄环境的环境参数进行确定，其中，环境参数可包括但不限于环境亮度、环境内容等，例如，根据环境亮度，拍摄场景可分为亮光场景、暗光场景、夜景场景、背光场景等，根据环境内容，拍摄场景可分为室内场景、天空场景、雨雾场景等，但不限于此。The electronic device can detect the current shooting scene, and the shooting scene can be determined according to the environmental parameters of the shooting environment, where the environmental parameters can include but are not limited to environmental brightness, environmental content, etc. For example, according to the environmental brightness, the shooting scene can be divided into bright scenes. , dark light scene, night scene scene, backlight scene, etc. According to the environment content, the shooting scene can be divided into indoor scene, sky scene, rain and fog scene, etc., but not limited to this.

在一些实施例中，电子设备可判断当前拍摄场景是否为目标场景。在本申请实施例中，该目标场景可指的是场景亮度低于阈值的场景，例如暗光场景、夜景场景等。在目标场景中拍摄得到的图像，由于环境亮度较低，会丢失较多的图像信息，导致图像的成像效果差，视觉效果不佳等问题。In some embodiments, the electronic device may determine whether the current shooting scene is the target scene. In this embodiment of the present application, the target scene may refer to a scene where the scene brightness is lower than a threshold, such as a dark light scene, a night scene scene, and the like. The image captured in the target scene will lose more image information due to the low ambient brightness, resulting in poor imaging effect and poor visual effect of the image.

作为一种具体实施方式，电子设备可通过光照度传感器、光线传感器等检测当前拍摄场景的场景亮度，并判断检测到的场景亮度是否低于阈值。若场景亮度低于阈值，则可确定当前拍摄场景为目标场景。As a specific implementation manner, the electronic device can detect the scene brightness of the current shooting scene through an illuminance sensor, a light sensor, etc., and determine whether the detected scene brightness is lower than a threshold value. If the scene brightness is lower than the threshold, it can be determined that the current shooting scene is the target scene.

作为另一种具体实施方式，电子设备还可利用HDR算法进行打分，可采用不同的曝光参数采集当前拍摄场景的图像后，利用HDR算法对采集的多帧图像进行融合，并对融合得到的高动态图像进行打分。可将HDR算法的打分结果与场景亮度进行结合，若场景亮度低于阈值，且亮动态图像的分数高于设定的分数线，则可确定当前拍摄场景为目标场景。As another specific implementation manner, the electronic device can also use the HDR algorithm to score, and can use different exposure parameters to collect images of the current shooting scene, use the HDR algorithm to fuse the collected multiple frames of images, and fuse the high-resolution images obtained by fusion. Dynamic images for scoring. The scoring result of the HDR algorithm can be combined with the brightness of the scene. If the brightness of the scene is lower than the threshold and the score of the bright dynamic image is higher than the set score line, the current shooting scene can be determined as the target scene.

作为另一种具体实施方式，电子设备可获取待采集图像，并将待采集图像输入到场景检测模型中，通过场景检测模型确定当前拍摄场景。场景检测模型可提取待采集图像的图像特征，并对图像特征进行分析，从而确定对应的拍摄场景。场景检测模型可根据大量携带有场景标签的场景图像样本进行训练构建，通过对大量携带有场景标签的场景图像样本进行学习，可以不断学习得到不同拍摄场景对应的图像特征，使得场景检测的结果更为准确。As another specific implementation manner, the electronic device may acquire the to-be-collected image, input the to-be-collected image into the scene detection model, and determine the current shooting scene through the scene detection model. The scene detection model can extract the image features of the image to be collected, and analyze the image features to determine the corresponding shooting scene. The scene detection model can be trained and constructed based on a large number of scene image samples with scene labels. By learning a large number of scene image samples with scene labels, the image features corresponding to different shooting scenes can be continuously learned, which makes the results of scene detection more accurate. to be accurate.

步骤504，当确定当前拍摄场景为目标场景时，对待采集图像进行人脸识别，目标场景的场景亮度低于阈值。Step 504 , when it is determined that the current shooting scene is the target scene, face recognition is performed on the image to be collected, and the scene brightness of the target scene is lower than the threshold.

若当前拍摄场景为目标场景，可说明当前拍摄场景的场景亮度较低，拍摄得到的图像的成像效果不佳，需对拍摄的图像进行处理，以提高图像的成像效果。电子设备在确定当前拍摄场景为目标场景后，可对待采集图像进行人脸识别，当检测到待采集图像中包含人脸时，可执行步骤506至512，得到成像效果高的目标图像。If the current shooting scene is the target scene, it means that the scene brightness of the current shooting scene is low, the imaging effect of the captured image is not good, and the captured image needs to be processed to improve the imaging effect of the image. After determining that the current shooting scene is the target scene, the electronic device can perform face recognition on the image to be collected, and when it is detected that the image to be collected contains a face,steps 506 to 512 can be performed to obtain a target image with high imaging effect.

人脸检测的具体内容，以及步骤506至512的内容可参照上述实施例中的相关描述，在此不再一一赘述。The specific content of face detection and the content ofsteps 506 to 512 can be referred to the relevant descriptions in the above embodiments, which will not be repeated here.

步骤506，当待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像。Step 506 , when the image to be collected includes a face region, collect multiple frames of first images according to multiple sets of exposure parameters.

在一些实施例中，曝光参数可包括曝光补偿值，利用曝光补偿值可增加或减少曝光量。当待采集图像中包含人脸区域时，可采用第一数值范围内的多个曝光补偿值采集得到多帧第一图像。例如，第一数值范围为[EV-3，EV+3]，若采集5帧第一图像，则可从该数值范围选取5个曝光补偿值，可分别为EV-3、EV-1、EV0、EV+2、EV+3。In some embodiments, the exposure parameter may include an exposure compensation value, with which the exposure amount may be increased or decreased. When the image to be collected includes a face region, multiple frames of the first image can be obtained by collecting multiple exposure compensation values within the first numerical range. For example, the first value range is [EV-3, EV+3]. If 5 frames of the first image are collected, 5 exposure compensation values can be selected from this value range, which can be EV-3, EV-1, EV0 respectively. , EV+2, EV+3.

若待采集图像中不包含人脸区域，则可采用第二数值范围内的多个曝光补偿值采集多帧第一图像，其中，第二数值范围可区别地第一数值范围。例如，第二数值范围为[EV-2，EV+2]，若采集5帧第一图像，则可从该数值范围选取5个曝光补偿值，可分别为EV-2、EV-1、EV0、EV+1、EV+2。可直接对采集的多帧第一图像进行融合处理，得到拥有高动态范围的目标图像。If the to-be-captured image does not include a face area, multiple exposure compensation values within a second numerical range may be used to collect multiple frames of the first image, wherein the second numerical range is distinguishable from the first numerical range. For example, the second value range is [EV-2, EV+2]. If 5 frames of the first image are collected, 5 exposure compensation values can be selected from this value range, which can be EV-2, EV-1, EV0 respectively. , EV+1, EV+2. The multi-frame first images collected can be directly fused to obtain a target image with a high dynamic range.

在一些实施方式中，第一数值范围及第二数值范围均可根据当前拍摄场景的场景亮度及感光度(ISO值)进行确定，感光度可用于衡量底片对于光的灵敏程度。若场景亮度较低且感光度较低，则设置的第一数值范围及第二数值范围可较宽，以保证能采集到具有充足曝光量的图像，若场景亮度较高且感光度较高，则设置的第一数值范围及第二数值范围可较窄，以保证采集到的图像拥有更丰富的图像信息。电子设备中可预先存储有场景亮度及感光度与第一数值范围、第二数值范围之间的对应关系，在确定场景亮度及感光度后，可根据该对应关系获取第一数值范围及第二数值范围，从而根据第一数值范围及第二数值范围采集多帧第一图像。In some embodiments, the first numerical range and the second numerical range can be determined according to the scene brightness and sensitivity (ISO value) of the current shooting scene, and the sensitivity can be used to measure the sensitivity of the negative to light. If the scene brightness is low and the sensitivity is low, the first value range and the second value range can be set wider to ensure that images with sufficient exposure can be captured. If the scene brightness is high and the sensitivity is high, Then, the set first numerical range and second numerical range may be narrower, so as to ensure that the collected image has richer image information. The electronic device may pre-store the correspondence between the scene brightness and sensitivity and the first numerical range and the second numerical range. After determining the scene brightness and sensitivity, the first numerical range and the second numerical range may be obtained according to the corresponding relationship. value range, so that multiple frames of first images are collected according to the first value range and the second value range.

步骤508，对多帧第一图像进行融合处理，得到第二图像。Step 508: Perform fusion processing on multiple frames of the first image to obtain a second image.

步骤510，提取第一图像中包含的人像区域，并对人像区域进行增强处理，人像区域包括人脸区域。Step 510: Extract the portrait area included in the first image, and perform enhancement processing on the portrait area, where the portrait area includes a human face area.

可从采集的多帧第一图像中选取至少一帧第一图像，并从选取的第一图像中提取人像区域。在一些实施方式中，可从采集的多帧第一图像中选取至少一帧曝光补偿值小于0的第一图像。在目标场景下，采用曝光补偿值大于或等于0拍摄得到的第一图像的图像噪声可能较大，即便使用降噪算法也可能无法完全去除图像噪声，因此，可选取曝光补偿值小于0的第一图像。进一步地，可仅选取一张曝光补偿值小于0的第一图像进行人物区域的技取。当存在多张曝光补偿值小于0的第一图像时，可从中随机选取一张第一图像，也可根据每张曝光补偿值小于0的第一图像的图像质量进行选取。At least one frame of the first image may be selected from the collected multiple frames of the first image, and a portrait area may be extracted from the selected first image. In some embodiments, at least one frame of the first image whose exposure compensation value is less than 0 may be selected from the collected multiple frames of the first image. In the target scene, the image noise of the first image obtained by using the exposure compensation value greater than or equal to 0 may be large, and even if the noise reduction algorithm is used, the image noise may not be completely removed. Therefore, the first image with the exposure compensation value less than 0 can be selected an image. Further, only one first image with an exposure compensation value of less than 0 may be selected to perform the technique of the person area. When there are multiple first images whose exposure compensation value is less than 0, one first image may be randomly selected from among them, or may be selected according to the image quality of each first image whose exposure compensation value is less than 0.

如图6所示，在一个实施例中，选取至少一帧曝光补偿值小于0的第一图像可包括以下步骤：As shown in FIG. 6 , in one embodiment, selecting at least one frame of the first image whose exposure compensation value is less than 0 may include the following steps:

步骤602，检测每帧曝光补偿值小于0的第一图像的图像噪声。Step 602: Detect image noise of the first image whose exposure compensation value per frame is less than 0.

图像噪声是指图像中存在的不必要的或多余的干扰信息。可选地，可采用噪声检测算法对每帧曝光补偿值小于0的第一图像进行检测，例如，可采用直方图统计、分区均值算法等检测图像噪声。Image noise refers to unnecessary or redundant interfering information existing in an image. Optionally, a noise detection algorithm may be used to detect the first image whose exposure compensation value of each frame is less than 0, for example, image noise may be detected by using histogram statistics, a partition mean algorithm, or the like.

在一些实施例中，还可将各帧曝光补偿值小于0的第一图像进行比对，从而确定每帧曝光补偿值小于0的第一图像的图像噪声。可进行检测的第一图像的像素点的灰度值与其他曝光补偿值小于0的第一图像匹配的像素点的灰度值进行比较，若检测的像素点与其他第一图像中匹配的像素点的灰度值差值大于波动范围，则可确定该检测的像素点为图像噪声。可以理解地，也可采用其他方式进行图像噪声检测，在此不进行限定。In some embodiments, the first images with the exposure compensation value less than 0 in each frame may also be compared, so as to determine the image noise of the first image with the exposure compensation value less than 0 in each frame. The gray value of the pixel point of the first image that can be detected is compared with the gray value of the matching pixel point of other first images whose exposure compensation value is less than 0. If the difference of the gray value of the point is greater than the fluctuation range, it can be determined that the detected pixel point is image noise. It can be understood that the image noise detection can also be performed in other manners, which are not limited herein.

步骤604，根据每帧曝光补偿值小于0的第一图像的图像噪声及包含的人像信息，确定每帧曝光补偿值小于0的第一图像的图像质量。Step 604: Determine the image quality of the first image with the exposure compensation value of each frame less than 0 according to the image noise and the included portrait information of the first image with the exposure compensation value of each frame.

人像信息可包括人像的细节信息、亮度信息、饱和度等，人像信息越丰富，可说明第一图像的图像质量越好。作为一种具体实施方式，可根据每帧曝光补偿值小于0的第一图像的图像噪声及包含的人像信息，计算每帧曝光补偿值小于0的第一图像的质量分数，该质量分数可用于表征图像质量。第一图像的图像噪声越少，包含的人像信息越丰富，对应的质量分数可越高，表示图像质量越高；第一图像的图像噪声越多，包含的人像信息越少，对应的质量分数可越低，表示图像质量越低。The portrait information may include detail information, brightness information, saturation, etc. of the portrait. The richer the portrait information, the better the image quality of the first image. As a specific implementation manner, the quality score of the first image with the exposure compensation value of each frame less than 0 can be calculated according to the image noise of the first image with the exposure compensation value of each frame less than 0 and the included portrait information, and the quality score can be used for Characterize image quality. The less image noise in the first image, the richer the portrait information it contains, and the higher the corresponding quality score, which means the higher the image quality; the more image noise in the first image, the less portrait information it contains, and the corresponding quality score The lower the value, the lower the image quality.

步骤606，基于图像质量选取一帧曝光补偿值小于0的第一图像。Step 606 , selecting a frame of a first image with an exposure compensation value less than 0 based on the image quality.

电子设备可基于每帧曝光补偿值小于0的第一图像的图像质量，按照图像质量从高到低进行排序，可选择图像质量最高的曝光补偿值小于0的第一图像，并提取该选取的第一图像包含的人像区域，再对该人像区域进行增强处理。可使得提取的人像区域包含的图像信息更多，进一步提高后续得到的目标图像的人像的视觉效果。此外，提取的人像区域包含的图像信息更多，可减少增强处理时的运算工作量，提高图像处理效率。The electronic device may, based on the image quality of the first image with the exposure compensation value of each frame less than 0, sort the image quality from high to low, select the first image with the exposure compensation value of the highest image quality less than 0, and extract the selected first image with the exposure compensation value less than 0. The portrait area included in the first image is then enhanced. The extracted portrait region can contain more image information, and further improve the visual effect of the portrait of the target image obtained subsequently. In addition, the extracted portrait region contains more image information, which can reduce the computational workload during enhancement processing and improve image processing efficiency.

步骤512，将增强处理后的人像区域与第二图像进行拼接，得到目标图像。Step 512 , splicing the enhanced portrait region with the second image to obtain a target image.

在本申请实施例中，通过检测当前拍摄场景，当当前拍摄场景为场景亮度低于阈值的目标场景时，对待采集图像进行人脸识别，根据多组曝光参数采集多帧第一图像，对该多帧第一图像进行融合处理，得到第二图像，提取第一图像中包含的人像区域，单独对人像区域进行增强处理后再与由多帧第一图像融合得到的第二图像拼接，能够保证在场景亮度低的情况下拍摄的整体图像既拥有更多的亮度范围，又能提高人物图像视觉效果。In the embodiment of the present application, by detecting the current shooting scene, when the current shooting scene is a target scene with a scene brightness lower than a threshold, face recognition is performed on the image to be collected, and multiple frames of first images are collected according to multiple sets of exposure parameters. Multi-frame first images are fused to obtain a second image, the portrait area contained in the first image is extracted, the portrait area is individually enhanced, and then spliced with the second image obtained by merging multiple frames of the first image, which can ensure The overall image captured in the case of low scene brightness has more brightness range and can improve the visual effect of human image.

如图7所示，在一个实施例中，提供另一种图像处理方法，可应用于上述的电子设备。该方法可包括如下步骤：As shown in FIG. 7 , in one embodiment, another image processing method is provided, which can be applied to the above-mentioned electronic device. The method may include the steps of:

步骤702，检测当前拍摄场景。Step 702, detecting the current shooting scene.

步骤704，当确定当前拍摄场景为目标场景时，对待采集图像进行人脸识别，目标场景的场景亮度低于阈值。Step 704 , when it is determined that the current shooting scene is the target scene, face recognition is performed on the image to be collected, and the scene brightness of the target scene is lower than the threshold.

步骤706，当待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像。Step 706: When the image to be collected includes a face region, collect multiple frames of first images according to multiple sets of exposure parameters.

步骤708，对多帧第一图像进行融合处理，得到第二图像。Step 708: Perform fusion processing on multiple frames of the first image to obtain a second image.

步骤710，提取第一图像中包含的人像区域，并对人像区域进行增强处理，人像区域包括人脸区域。Step 710 , extracting a portrait area included in the first image, and performing enhancement processing on the portrait area, where the portrait area includes a human face area.

步骤702至710的内容可参照上述实施例中的相关描述，在此不再进行一一赘述。For the content ofsteps 702 to 710, reference may be made to the relevant descriptions in the above-mentioned embodiments, which will not be repeated here.

在一些实施方式中，对人像区域进行增强处理，可包括提升人像区域的清晰度、消除人像区域中存在的模糊区域、根据亮度范围调整人像区域中人脸区域的亮度、确定人像区域的皮肤区域，并提高皮肤区域的饱和度、增强人像区域中包含的五官区域的细节信息等中的一种或多种等。In some embodiments, performing enhancement processing on the portrait area may include improving the clarity of the portrait area, eliminating blurred areas in the portrait area, adjusting the brightness of the face area in the portrait area according to the brightness range, and determining the skin area of the portrait area. , and increase the saturation of the skin area, enhance the detail information of the facial features area included in the portrait area, and so on.

作为一种具体实施方式，可提高人像区域的分辨率，也可以对人像区域中的像素点进行锐化处理，从而可提升人像区域的清晰度。As a specific implementation manner, the resolution of the portrait area can be improved, and the pixel points in the portrait area can also be sharpened, so that the definition of the portrait area can be improved.

作为一种具体实施方式，可检测人像区域中存在的模糊区域，并对该模糊区域进行锐化处理，从而消除人像区域中存在的模糊区域。可选地，也可获取其它第一图像中与该模糊区域对应的图像区域，并选取清晰度最高的图像区域与该人像区域中的模糊区域融合，消除人像区域中存在的模糊区域。As a specific implementation manner, a blurred area existing in the portrait area may be detected, and a sharpening process may be performed on the blurred area, thereby eliminating the blurred area existing in the portrait area. Optionally, the image area corresponding to the blurred area in other first images may also be acquired, and the image area with the highest definition is selected to be fused with the blurred area in the portrait area to eliminate the blurred area existing in the portrait area.

作为一种具体实施方式，亮度范围可根据实际需求进行设定，根据亮度范围调整人像区域中人脸区域的亮度，可使人脸区域的亮度不会过亮或过暗。As a specific implementation manner, the brightness range can be set according to actual requirements, and the brightness of the face area in the portrait area is adjusted according to the brightness range, so that the brightness of the face area will not be too bright or too dark.

饱和度可定义为彩度除以明度，可用于表征彩色偏离同亮度灰色的程度。可提高人像区域的皮肤区域的饱和度，以提高人像区域的视觉效果。作为一种具体实施方式，可先检测人像区域的皮肤区域，可先将人像区域划分为多个区域，再判断各个区域是否与预设的皮肤区域的特征匹配，若特征匹配，则可确定为皮肤区域，再提高皮肤区域的饱和度。Saturation can be defined as chroma divided by lightness and can be used to characterize how much a color deviates from a gray of the same brightness. The saturation of the skin area of the portrait area can be increased to improve the visual effect of the portrait area. As a specific implementation, the skin area of the portrait area can be detected first, the portrait area can be divided into multiple areas, and then it is judged whether each area matches the characteristics of the preset skin area, and if the characteristics match, it can be determined as The skin area, and then increase the saturation of the skin area.

作为一种具体实施方式，可以增强人像区域中五官区域的五官轮廓，突出五官的边缘信息，对五官区域进行增强处理。As a specific implementation manner, the outline of the facial features of the facial features region in the portrait region can be enhanced, the edge information of the facial features can be highlighted, and the facial feature region can be enhanced.

可以理解地，本申请实施例中还可对人像区域进行比上述所举的几种更多的增强处理，也可以对人像区域进行美化处理，例如瘦身、美白、拉长身体比例等，在此不进行一一限定。It can be understood that, in the embodiments of the present application, more enhancement processing can be performed on the portrait area than those mentioned above, and beautification processing can also be performed on the portrait area, such as slimming, whitening, and lengthening the proportion of the body, etc. Not limited to one by one.

步骤712，确定第二图像中与人像区域对应的子区域。Step 712: Determine a sub-region corresponding to the portrait region in the second image.

对提取的人像区域进行增强处理后，可将增强处理的人像区域与第二图像进行亮度对齐，亮度对齐指的是使人像区域的亮度与第二图像的亮度相同或相似。在一些实施例中，可将增强处理后的人像区域与第二图像中对应的子区域进行亮度对齐。After the extracted portrait region is enhanced, the enhanced portrait region can be brightness aligned with the second image. Brightness alignment refers to making the brightness of the portrait region the same as or similar to the brightness of the second image. In some embodiments, the enhanced portrait area may be luminance aligned with the corresponding sub-areas in the second image.

可确定第二图像中与人像区域对应的子区域，作为一种实施方式，可获取人像区域的各个像素点在所属的第一图像的像素坐标，并将第二图像中与该所属的第一图像的像素坐标相同的像素点确定为子区域的像素点，从而可确定第二图像中与人像区域对应的子区域。The sub-region corresponding to the portrait region in the second image can be determined. As an implementation manner, the pixel coordinates of each pixel in the portrait region in the first image to which it belongs can be obtained, and the second image can be compared with the first image to which it belongs. The pixel points with the same pixel coordinates of the image are determined as the pixel points of the sub-region, so that the sub-region corresponding to the portrait region in the second image can be determined.

作为另一种实施方式，也可直接对第二图像进行人像区域检测，其检测的方式可与上述实施例中从第一图像中提取人像区域时的检测方式一致，在此不再一一赘述。可将第二图像中检测到的人像区域作为与增强处理后的人像区域对应的子区域。As another implementation, it is also possible to directly perform the detection of the portrait region on the second image, and the detection method can be the same as the detection method when the portrait region is extracted from the first image in the above-mentioned embodiment, and will not be repeated here. . The portrait region detected in the second image may be used as a sub-region corresponding to the enhanced portrait region.

作为另一种实施方式，还可将增强处理后的人像区域与第二图像进行匹配，确定第二图像中与人像区域的边缘像素点匹配的像素点，从而划分出与该人像区域对应的子区域。As another implementation manner, the enhanced portrait region can also be matched with the second image, and the pixels in the second image that match the edge pixels of the portrait region can be determined, so as to divide the sub-portraits corresponding to the portrait region. area.

步骤714，根据子区域的亮度调整增强处理后的人像区域的亮度，使得增强处理后的人像区域的亮度向子区域的亮度对齐。Step 714: Adjust the brightness of the enhanced portrait region according to the brightness of the subregion, so that the brightness of the enhanced portrait region is aligned with the brightness of the subregion.

可通过多种方式将增强处理后的人像区域的亮度与第二图像中对应的子区域进行亮度对齐，例如，可通过直方图分布方法进行调整，也可以通过计算平均亮度值进行调整等。The brightness of the enhanced portrait region can be aligned with the corresponding sub-regions in the second image in various ways. For example, it can be adjusted by a histogram distribution method, or by calculating an average brightness value.

在一些实施例中，可计算第二图像中与增强处理后的人像区域对应的子区域的第一平均亮度，并计算增强处理后的人像区域的第二平均亮度。可比较该第一平均亮度及第二平均亮度，得到比较结果，并根据比较结果确定调整参数。若比较结果为第一平均亮度大于第二平均亮度，则可提高增强处理后的人像区域的亮度。若比较结果为第一平均亮度小于第二平均亮度，则可降低增强处理后的人像区域的亮度。In some embodiments, the first average brightness of the sub-regions corresponding to the enhanced portrait region in the second image may be calculated, and the second average brightness of the enhanced portrait region may be calculated. The first average brightness and the second average brightness can be compared to obtain a comparison result, and the adjustment parameter can be determined according to the comparison result. If the comparison result is that the first average brightness is greater than the second average brightness, the brightness of the enhanced portrait area can be increased. If the comparison result is that the first average brightness is smaller than the second average brightness, the brightness of the enhanced-processed portrait area may be reduced.

调整参数可根据第一平均亮度与第二平均亮度之间的差值进行确定，以保证进行亮度调整后的人脸区域的第二平均亮度与第二图像对应的子区域的第一平均亮度的差值满足设定的误差范围。可选地，误差范围可根据实际需求进行设定，通常为较小的数值范围，比如为[0，3]等，以保证亮度调整后的人脸区域的亮度与第二图像中对应子区域的亮度相同或相似，实现亮度对齐，可提高拼接后的图像的视觉效果。The adjustment parameter can be determined according to the difference between the first average brightness and the second average brightness, so as to ensure the difference between the second average brightness of the face region after brightness adjustment and the first average brightness of the sub-region corresponding to the second image. The difference satisfies the set error range. Optionally, the error range can be set according to actual needs, and is usually a small value range, such as [0, 3], etc., to ensure that the brightness of the face region after brightness adjustment is the same as the corresponding sub-region in the second image. The brightness of the images is the same or similar, and the brightness alignment can be achieved, which can improve the visual effect of the spliced image.

步骤716，将增强处理后的人像区域与第二图像进行拼接，得到目标图像。Step 716, splicing the enhanced portrait region with the second image to obtain a target image.

可将进行亮度对齐后的增强处理后的人像区域与第二图像进行拼接，得到目标图像，具体的实施方式可参照上述实施例中的描述，在此不再一一赘述。The target image can be obtained by splicing the enhanced portrait region after brightness alignment with the second image, and the specific implementation can refer to the description in the above-mentioned embodiment, which will not be repeated here.

在一些实施方式中，电子设备将增强处理后的人像区域与第二图像进行拼接后，还可对增强处理后的人像区域与第二图像的拼接边缘进行平滑处理。可选地，可采用均值滤波、高斯加权滤波、中值滤波、双边滤波等方式对拼接边缘进行平滑处理，利用拼接边缘像素点的相邻像素点对拼接边缘像素点进行滤波，减小拼接边缘像素点与相邻像素点之间的差异，实现平滑处理，可使得拼接后的目标图像更为自然，提高成像效果。In some embodiments, after the electronic device splices the enhanced portrait region and the second image, the electronic device may further perform smoothing processing on the spliced edge of the enhanced portrait region and the second image. Optionally, mean filtering, Gaussian weighted filtering, median filtering, bilateral filtering, etc. can be used to smooth the splicing edge, and the adjacent pixels of the splicing edge pixel are used to filter the splicing edge pixels to reduce the splicing edge. The difference between the pixel point and the adjacent pixel point can be smoothed, which can make the spliced target image more natural and improve the imaging effect.

在本申请实施例中，可将增强处理后的人像区域与第二图像进行亮度对齐，保证二者亮度相同或相似，可提高拼接后的目标图像的视角效果。In the embodiment of the present application, the brightness of the enhanced portrait region and the second image can be aligned to ensure that the brightness of the two is the same or similar, which can improve the viewing angle effect of the spliced target image.

图8为另一个实施例中图像处理方法的流程示意图。如图8所示，在一个实施例中，可利用多组不同的曝光补偿值(包括EV+、EV0及EV-等)采集多帧曝光量不同的第一图像，并对采集的多帧第一图像进行RAW域处理，以对多帧第一图像进行降噪。降噪后的多帧第一图像可输入到ISP处理器中，由ISP处理器进行处理。ISP处理器可采用HDR算法对多帧第一图像进行融合，得到具有高动态范围的第二图像。ISP处理器还可选取一帧曝光补偿值为EV-的第一图像进行人像分割，提取人像区域，并对人像区域进行细节增强等增强处理。可将增强处理后的人像区域与第二图像进行亮度对齐，并将亮度对齐后的增强处理后的人像区域与第二图像进行拼接。可对增强处理后的人像区域与第二图像的拼接边缘进行平滑处理，最终得到目标图像。在保证目标图像的拥有高动态范围的同时，也增强人像的视觉效果，能够使人物图像展示更多的图像细节，提高人物图像的成像质量。FIG. 8 is a schematic flowchart of an image processing method in another embodiment. As shown in FIG. 8 , in one embodiment, multiple sets of different exposure compensation values (including EV+, EV0, EV-, etc.) may be used to collect multiple frames of first images with different exposures, and the collected multiple frames of first images The images are processed in the RAW domain to denoise the first image of multiple frames. The multi-frame first image after noise reduction can be input into the ISP processor and processed by the ISP processor. The ISP processor may use the HDR algorithm to fuse multiple frames of the first image to obtain a second image with a high dynamic range. The ISP processor can also select a frame of the first image with the exposure compensation value of EV- to perform portrait segmentation, extract the portrait area, and perform enhancement processing such as detail enhancement on the portrait area. The enhanced portrait area and the second image may be brightness aligned, and the enhanced portrait area after brightness alignment may be stitched with the second image. The enhanced portrait area and the splicing edge of the second image can be smoothed to finally obtain the target image. While ensuring the high dynamic range of the target image, it also enhances the visual effect of the portrait, which enables the human image to display more image details and improves the imaging quality of the human image.

如图9所示，在一个实施例中，提供一种图像处理装置900，包括人脸识别模块910、采集模块920、融合模块930、增强模块940及拼接模块950。As shown in FIG. 9 , in one embodiment, animage processing apparatus 900 is provided, including aface recognition module 910 , acollection module 920 , afusion module 930 , anenhancement module 940 and astitching module 950 .

人脸识别模块910，用于对待采集图像进行人脸识别。Theface recognition module 910 is used for performing face recognition on the image to be collected.

采集模块920，用于当待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像。Thecollection module 920 is configured to collect multiple frames of first images according to multiple sets of exposure parameters when the image to be collected includes a face region.

融合模块930，用于对多帧第一图像进行融合处理，得到第二图像。Thefusion module 930 is configured to perform fusion processing on multiple frames of the first image to obtain a second image.

增强模块940，用于提取第一图像中包含的人像区域，并对人像区域进行增强处理，人像区域包括人脸区域。Theenhancement module 940 is configured to extract the portrait region included in the first image, and perform enhancement processing on the portrait region, where the portrait region includes the human face region.

拼接模块950，用于将增强处理后的人像区域与第二图像进行拼接，得到目标图像。Thesplicing module 950 is used for splicing the enhanced portrait region and the second image to obtain the target image.

在本申请实施例中，通过对待采集图像进行人脸识别，当待采集图像中包含人脸区域时，根据多组曝光参数采集多帧第一图像，对该多帧第一图像进行融合处理，得到第二图像，提取第一图像中包含的人像区域，并对人像区域进行增强处理，再将增强处理后的人像区域与第二图像进行拼接，得到目标图像，单独对人像区域进行增强处理后再与由多帧第一图像融合得到的第二图像拼接，在保证整体图像的成像效果的同时，也增强人像的视觉效果，能够使人物图像展示更多的图像细节，提高人物图像的成像质量。In the embodiment of the present application, by performing face recognition on the image to be collected, when the image to be collected includes a face area, collect multiple frames of first images according to multiple sets of exposure parameters, and perform fusion processing on the multiple frames of first images, Obtain the second image, extract the portrait area contained in the first image, perform enhancement processing on the portrait area, and then splicing the enhanced portrait area with the second image to obtain the target image, and enhance the portrait area separately. It is then spliced with the second image obtained by merging multiple frames of the first image, which not only ensures the imaging effect of the overall image, but also enhances the visual effect of the portrait, which can make the portrait image show more image details and improve the imaging quality of the portrait image. .

在一个实施例中，上述图像处理装置900，除了包括人脸识别模块910、采集模块920、融合模块930、增强模块940及拼接模块950，还包括场景检测模块。In one embodiment, the aboveimage processing apparatus 900, in addition to theface recognition module 910, theacquisition module 920, thefusion module 930, theenhancement module 940, and thestitching module 950, also includes a scene detection module.

场景检测模块，用于检测当前拍摄场景。The scene detection module is used to detect the current shooting scene.

人脸识别模块910，还用于当确定当前拍摄场景为目标场景时，对待采集图像进行人脸识别，目标场景的场景亮度低于阈值。Theface recognition module 910 is further configured to perform face recognition on the image to be collected when it is determined that the current shooting scene is the target scene, and the scene brightness of the target scene is lower than the threshold.

在一个实施例中，曝光参数包括曝光补偿值。In one embodiment, the exposure parameters include exposure compensation values.

增强模块940，还用于从多帧第一图像中选取至少一帧曝光补偿值小于0的第一图像，并提取选取的第一图像中包含的人像区域。Theenhancement module 940 is further configured to select at least one first image whose exposure compensation value is less than 0 from the multiple frames of the first images, and extract the portrait area included in the selected first image.

在一个实施例中，增强模块940，包括噪声检测单元、质量确定单元及选取单元。In one embodiment, theenhancement module 940 includes a noise detection unit, a quality determination unit and a selection unit.

噪声检测单元，用于检测每帧曝光补偿值小于0的第一图像的图像噪声。A noise detection unit, configured to detect image noise of the first image whose exposure compensation value of each frame is less than 0.

质量确定单元，用于根据每帧曝光补偿值小于0的第一图像的图像噪声及包含的人像信息，确定每帧曝光补偿值小于0的第一图像的图像质量。The quality determination unit is configured to determine the image quality of the first image with the exposure compensation value of each frame less than 0 according to the image noise and the included portrait information of the first image with the exposure compensation value of each frame.

选取单元，用于基于图像质量选取一帧曝光补偿值小于0的第一图像。A selection unit, configured to select a frame of a first image with an exposure compensation value less than 0 based on image quality.

在本申请实施例中，通过检测当前拍摄场景，当当前拍摄场景为场景亮度低于阈值的目标场景时，对待采集图像进行人脸识别，根据多组曝光参数采集多帧第一图像，对该多帧第一图像进行融合处理，得到第二图像，提取第一图像中包含的人像区域，单独对人像区域进行增强处理后再与由多帧第一图像融合得到的第二图像拼接，能够保证在场景亮度低的情况下拍摄的整体图像既拥有更多的亮度范围，又能提高人物图像视觉效果。In the embodiment of the present application, by detecting the current shooting scene, when the current shooting scene is a target scene with a scene brightness lower than a threshold, face recognition is performed on the image to be collected, and multiple frames of first images are collected according to multiple sets of exposure parameters. Multi-frame first images are fused to obtain a second image, the portrait area contained in the first image is extracted, the portrait area is individually enhanced, and then spliced with the second image obtained by merging multiple frames of the first image, which can ensure The overall image captured in the case of low scene brightness has more brightness range and can improve the visual effect of human image.

在一个实施例中，上述图像处理装置900，除了包括人脸识别模块910、采集模块920、融合模块930、增强模块940、拼接模块950及场景检测模块，还包括子区域确定模块及亮度对齐模块。In one embodiment, the aboveimage processing apparatus 900, in addition to theface recognition module 910, theacquisition module 920, thefusion module 930, theenhancement module 940, thesplicing module 950 and the scene detection module, also includes a sub-region determination module and a brightness alignment module .

子区域确定模块，用于确定第二图像中与人像区域对应的子区域。The sub-region determining module is configured to determine the sub-region corresponding to the portrait region in the second image.

亮度对齐模块，用于根据子区域的亮度调整增强处理后的人像区域的亮度，使得增强处理后的人像区域的亮度向子区域的亮度对齐。The brightness alignment module is configured to adjust the brightness of the enhanced portrait area according to the brightness of the sub area, so that the brightness of the enhanced portrait area is aligned with the brightness of the sub area.

在一个实施例中，亮度对齐模块，包括计算单元、比较单元及调整单元。In one embodiment, the luminance alignment module includes a calculation unit, a comparison unit and an adjustment unit.

计算单元，用于计算子区域的第一平均亮度及增强处理后的人像区域的第二平均亮度。The calculating unit is configured to calculate the first average brightness of the sub-region and the second average brightness of the enhanced portrait region.

比较单元，用于比较第一平均亮度及第二平均亮度，得到比较结果，并根据比较结果确定调整参数。The comparison unit is used for comparing the first average brightness and the second average brightness, obtaining a comparison result, and determining an adjustment parameter according to the comparison result.

调整单元，用于根据调整参数调整增强处理后的人像区域的亮度，使得亮度调整后的人脸区域的第二平均亮度与第一平均亮度的差值满足误差范围。The adjustment unit is configured to adjust the brightness of the enhanced portrait region according to the adjustment parameters, so that the difference between the second average brightness and the first average brightness of the brightness-adjusted human face region satisfies the error range.

在一个实施例中，增强模块940，还用于对人像区域进行以下一种或多种的增强处理：In one embodiment, theenhancement module 940 is further configured to perform one or more of the following enhancement processing on the portrait area:

提升人像区域的清晰度；Improve the clarity of the portrait area;

消除人像区域中存在的模糊区域；Eliminate the blurred areas that exist in the portrait area;

根据亮度范围调整人像区域中人脸区域的亮度；Adjust the brightness of the face area in the portrait area according to the brightness range;

确定人像区域的皮肤区域，并提高皮肤区域的饱和度；Determine the skin area of the portrait area and increase the saturation of the skin area;

增强人像区域中包含的五官区域的细节信息。Enhance the details of the facial features included in the portrait area.

在一个实施例中，拼接模块950，包括拼接单元及平滑单元。In one embodiment, thesplicing module 950 includes a splicing unit and a smoothing unit.

拼接单元，用于将增强处理后的人像区域与第二图像进行拼接。The splicing unit is used for splicing the enhanced portrait area with the second image.

平滑单元，用于对增强处理后的人像区域与第二图像的拼接边缘进行平滑处理，得到目标图像。The smoothing unit is used for smoothing the spliced edge of the enhanced portrait region and the second image to obtain the target image.

在本申请实施例中，可将增强处理后的人像区域与第二图像进行亮度对齐，保证二者亮度相同或相似，可提高拼接后的目标图像的视角效果。In the embodiment of the present application, the brightness of the enhanced portrait region and the second image can be aligned to ensure that the brightness of the two is the same or similar, which can improve the viewing angle effect of the spliced target image.

图10为一个实施例中电子设备的结构框图。如图10所示，电子设备1000可以包括一个或多个如下部件：处理器1010、与处理器1010耦合的存储器1020，其中存储器1020可存储有一个或多个应用程序，一个或多个应用程序可以被配置为由一个或多个处理器1010执行，一个或多个程序配置用于执行如上述实施例描述的方法。FIG. 10 is a structural block diagram of an electronic device in one embodiment. As shown in FIG. 10, theelectronic device 1000 may include one or more of the following components: aprocessor 1010, amemory 1020 coupled with theprocessor 1010, wherein thememory 1020 may store one or more application programs, one or more application programs The one or more programs may be configured to be executed by one ormore processors 1010 configured to perform the methods as described in the above embodiments.

处理器1010可以包括一个或者多个处理核。处理器1010利用各种接口和线路连接整个电子设备1000内的各个部分，通过运行或执行存储在存储器1020内的指令、程序、代码集或指令集，以及调用存储在存储器1020内的数据，执行电子设备1000的各种功能和处理数据。可选地，处理器1010可以采用数字信号处理(Digital Signal Processing，DSP)、现场可编程门阵列(Field－Programmable Gate Array，FPGA)、可编程逻辑阵列(Programmable Logic Array，PLA)中的至少一种硬件形式来实现。处理器1010可集成中央处理器(Central Processing Unit，CPU)、图像处理器(Graphics Processing Unit，GPU)和调制解调器等中的一种或几种的组合。其中，CPU主要处理操作系统、用户界面和应用程序等；GPU用于负责显示内容的渲染和绘制；调制解调器用于处理无线通信。可以理解的是，上述调制解调器也可以不集成到处理器1010中，单独通过一块通信芯片进行实现。Processor 1010 may include one or more processing cores. Theprocessor 1010 uses various interfaces and lines to connect various parts of the entireelectronic device 1000, and executes by running or executing the instructions, programs, code sets or instruction sets stored in thememory 1020, and calling the data stored in thememory 1020. Various functions of theelectronic device 1000 and processing data. Optionally, theprocessor 1010 may employ at least one of a digital signal processing (Digital Signal Processing, DSP), a Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA), and a Programmable Logic Array (Programmable Logic Array, PLA). A hardware form is implemented. Theprocessor 1010 may integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU), a modem, and the like. Among them, the CPU mainly handles the operating system, user interface and application programs, etc.; the GPU is used for rendering and drawing of the display content; the modem is used to handle wireless communication. It can be understood that, the above-mentioned modem may not be integrated into theprocessor 1010, and is implemented by a communication chip alone.

存储器1020可以包括随机存储器(Random Access Memory，RAM)，也可以包括只读存储器(Read-Only Memory)。存储器1020可用于存储指令、程序、代码、代码集或指令集。存储器1020可包括存储程序区和存储数据区，其中，存储程序区可存储用于实现操作系统的指令、用于实现至少一个功能的指令(比如触控功能、声音播放功能、图像播放功能等)、用于实现上述各个方法实施例的指令等。存储数据区还可以存储电子设备1000在使用中所创建的数据等。Thememory 1020 may include random access memory (Random Access Memory, RAM), or may include read-only memory (Read-Only Memory).Memory 1020 may be used to store instructions, programs, codes, sets of codes, or sets of instructions. Thememory 1020 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.) , instructions for implementing the foregoing method embodiments, and the like. The storage data area may also store data and the like created by theelectronic device 1000 in use.

可以理解地，电子设备1000可包括比上述结构框图中更多或更少的结构元件，例如，包括电源、输入按键、摄像头、扬声器、屏幕、RF(Radio Frequency，射频)电路、Wi-Fi(Wireless Fidelity，无线保真)模块、蓝牙模块、传感器等，还可在此不进行限定。It can be understood that theelectronic device 1000 may include more or less structural elements than those in the above structural block diagram, for example, including a power supply, an input button, a camera, a speaker, a screen, an RF (Radio Frequency, radio frequency) circuit, Wi-Fi ( Wireless Fidelity, wireless fidelity) modules, Bluetooth modules, sensors, etc., may not be limited here.

本申请实施例公开一种计算机可读存储介质，其存储计算机程序，其中，该计算机程序被处理器执行时实现如上述实施例描述的方法。The embodiment of the present application discloses a computer-readable storage medium, which stores a computer program, wherein, when the computer program is executed by a processor, the method described in the foregoing embodiments is implemented.

本申请实施例公开一种计算机程序产品，该计算机程序产品包括存储了计算机程序的非瞬时性计算机可读存储介质，且该计算机程序可被处理器执行时实现如上述实施例描述的方法。The embodiments of the present application disclose a computer program product, the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program can be executed by a processor to implement the methods described in the above embodiments.

本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一非易失性计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)等。Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium , when the program is executed, it may include the flow of the above-mentioned method embodiments. The storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), or the like.

如此处所使用的对存储器、存储、数据库或其它介质的任何引用可包括非易失性和/或易失性存储器。合适的非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)，它用作外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDR SDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)。Any reference to a memory, storage, database or other medium as used herein may include non-volatile and/or volatile memory. Suitable nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which acts as external cache memory. By way of illustration and not limitation, RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), Memory Bus (Rambus) Direct RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Memory Bus Dynamic RAM (RDRAM).

应理解，说明书通篇中提到的“一个实施例”或“一实施例”意味着与实施例有关的特定特征、结构或特性包括在本申请的至少一个实施例中。因此，在整个说明书各处出现的“在一个实施例中”或“在一实施例中”未必一定指相同的实施例。此外，这些特定特征、结构或特性可以以任意适合的方式结合在一个或多个实施例中。本领域技术人员也应该知悉，说明书中所描述的实施例均属于可选实施例，所涉及的动作和模块并不一定是本申请所必须的。It is to be understood that reference throughout the specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic associated with the embodiment is included in at least one embodiment of the present application. Thus, appearances of "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily necessarily referring to the same embodiment. Furthermore, the specific features, structures or characteristics may be combined in any suitable manner in one or more embodiments. Those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

在本申请的各种实施例中，应理解，上述各过程的序号的大小并不意味着执行顺序的必然先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。In the various embodiments of the present application, it should be understood that the size of the sequence numbers of the above-mentioned processes does not imply an inevitable sequence of execution, and the execution sequence of each process should be determined by its functions and internal logic, and should not be implemented in the present application. The implementation of the examples constitutes no limitation.

上述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物单元，即可位于一个地方，或者也可以分布到多个网络单元上。可根据实际的需要选择其中的部分或全部单元来实现本实施例方案的目的。The units described above as separate components may or may not be physically separated, and components displayed as units may or may not be object units, and may be located in one place or distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

另外，在本申请各实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

上述集成的单元若以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可获取的存储器中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或者部分，可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储器中，包括若干请求用以使得一台计算机设备(可以为个人计算机、服务器或者网络设备等，具体可以是计算机设备中的处理器)执行本申请的各个实施例上述方法的部分或全部步骤。The above-mentioned integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the whole or part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a memory , including several requests to cause a computer device (which may be a personal computer, a server, or a network device, etc., specifically a processor in the computer device) to execute some or all of the steps of the above methods in the various embodiments of the present application.

以上对本申请实施例公开的一种图像处理方法、装置、电子设备及计算机可读存储介质进行了详细介绍，本文中应用了具体个例对本申请的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本申请的方法及其核心思想。同时，对于本领域的一般技术人员，依据本申请的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本申请的限制。The image processing method, device, electronic device, and computer-readable storage medium disclosed in the embodiments of the present application have been described in detail above. The principles and implementations of the present application are described with specific examples. The descriptions are only used to aid in understanding the methodology of the present application and its core ideas. At the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific embodiments and application scope. To sum up, the content of this specification should not be construed as a limitation to the present application.

Claims (11)

1. An image processing method, comprising:

carrying out face recognition on an image to be collected;

when the image to be collected comprises a face area, collecting a plurality of frames of first images according to a plurality of groups of exposure parameters;

performing fusion processing on the multiple frames of first images to obtain a second image;

extracting a portrait area contained in the first image, and performing enhancement processing on the portrait area, wherein the portrait area comprises the face area;

and splicing the enhanced portrait area with the second image to obtain a target image.

2. The method according to claim 1, wherein before the face recognition of the image to be captured, the method further comprises:

detecting a current shooting scene;

the face recognition of the image to be collected comprises the following steps:

and when the current shooting scene is determined to be a target scene, carrying out face recognition on the image to be collected, wherein the scene brightness of the target scene is lower than a threshold value.

3. The method of claim 1, wherein the exposure parameter comprises an exposure compensation value;

the extracting of the portrait area included in the first image includes:

and selecting at least one first image with an exposure compensation value less than 0 from the plurality of first images, and extracting a portrait area contained in the selected first image.

4. The method according to claim 3, wherein said selecting at least one first image with an exposure compensation value less than 0 from said plurality of first images comprises:

detecting image noise of a first image with each frame of exposure compensation value smaller than 0;

determining the image quality of each frame of the first image with the exposure compensation value smaller than 0 according to the image noise of each frame of the first image with the exposure compensation value smaller than 0 and the contained portrait information;

and selecting a first image with an exposure compensation value less than 0 based on the image quality.

5. The method of claim 1, wherein prior to said stitching the enhanced portrait area with the second image, the method further comprises:

determining a sub-region corresponding to the portrait region in the second image;

and adjusting the brightness of the human image area after the enhancement processing according to the brightness of the sub-area, so that the brightness of the human image area after the enhancement processing is aligned to the brightness of the sub-area.

6. The method of claim 5, wherein the adjusting the brightness of the enhanced portrait area according to the brightness of the sub-area comprises:

calculating first average brightness of the sub-region and second average brightness of the enhanced portrait region;

comparing the first average brightness with the second average brightness to obtain a comparison result, and determining an adjustment parameter according to the comparison result;

and adjusting the brightness of the enhanced human image area according to the adjustment parameter, so that the difference value between the second average brightness and the first average brightness of the human face area after brightness adjustment meets the error range.

7. The method according to any one of claims 1 to 6, wherein the enhancing the portrait area comprises one or more of:

improving the definition of the portrait area;

eliminating a fuzzy area existing in the portrait area;

adjusting the brightness of the face area in the portrait area according to the brightness range;

determining a skin area of the portrait area and improving the saturation of the skin area;

enhancing detail information of a five sense organ region contained in the portrait region.

8. The method according to any one of claims 1 to 6, wherein the stitching the enhanced portrait area with the second image to obtain the target image comprises:

splicing the enhanced portrait area with the second image;

and smoothing the spliced edge of the portrait area after the enhancement processing and the second image to obtain a target image.

9. An image processing apparatus characterized by comprising:

the face recognition module is used for carrying out face recognition on the image to be collected;

the acquisition module is used for acquiring a plurality of frames of first images according to a plurality of groups of exposure parameters when the images to be acquired contain the face regions;

the fusion module is used for carrying out fusion processing on the multiple frames of first images to obtain a second image;

the enhancement module is used for extracting a portrait area contained in the first image and enhancing the portrait area, wherein the portrait area comprises the face area;

and the splicing module is used for splicing the enhanced portrait area with the second image to obtain a target image.

10. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program that, when executed by the processor, causes the processor to carry out the method of any one of claims 1 to 8.

11. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1 to 8.

Images